Enhancing OSA Assessment with Explainable AI.

Explainable Artificial Intelligence (xAI) is a rapidly growing field that focuses on making deep learning models interpretable and understandable to human decision-makers. In this study, we introduce xAAEnet, a novel xAI model applied to the assessment of Obstructive Sleep Apnea (OSA) severity. OSA is a prevalent sleep disorder that can lead to numerous medical conditions and is currently assessed using the Apnea-Hypopnea Index (AHI). However, AHI has been criticized for its inability to accurately estimate the effect of OSAs on related medical conditions. To address this issue, we propose a human-centric xAI approach that emphasizes similarity between apneic events as a whole and reduces subjectivity in diagnosis by examining how the model makes its decisions. Our model was trained and tested on a dataset of 60 patients' Polysomnographic (PSG) recordings. Our results demonstrate that the proposed model, xAAEnet, outperforms models with traditional architectures such as convolutional regressor, autoencoder (AE), and variational autoencoder (VAE). This study highlights the potential of xAI in providing an objective OSA severity scoring method.Clinical relevance- This study provides an objective OSA severity scoring technique which could improve the management of apneic patients in clinical practice.

The role of semantic assessment in the differential diagnosis between late-life depression and Alzheimer's disease or amnestic mild cognitive impairment: systematic review and meta-analysis.

OBJECT: The cognitive complaints encountered in late-life depression (LLD) make it difficult to distinguish from amnestic mild cognitive impairment (aMCI) and Alzheimer's disease (AD) based on an analysis of neurocognitive disorders. The hypothesis of the early impairment of semantic memory in AD and aMCI is considered a potential differential cognitive clue, but the absence of this impairment has not yet been confirmed in  LLD. METHOD: Based on the PRISMA method, we systematically seek neuropsychological assessments of individuals with LLD, the present study included 31 studies representing 3291 controls and 2820 people with LLD. Wherever possible, studies that tested simultaneously groups with LLD, AD (or aMCI) were also included. The results of the group of neuropsychological tasks relying on semantic memory were analyzed in two groups of tasks with high- or low-executive demand. The mean average effect of LLD was calculated and compared to the incremental effect of aMCI or AD on the scores. Linear regressions including education, age, and severity and type of depression were run to seek their power of prediction for the mean average effects. RESULTS: LLD has a medium effect on scores at semantic and phonemic fluency and naming and a small average effect on the low-executive demand tasks. Differences in education is a predictor of the effect of LLD on phonemic fluency and naming but not on semantic fluency or on low-executive demand tasks. Except for semantic fluency, aMCI did not demonstrate an incremental effect on the scores compared to LLD, while AD did, for all the tasks except phonemic fluency. CONCLUSION: Assessment of semantic memory can be a discriminating clue for the distinction between depression and Alzheimer's disease but some methodological variables are highly influential to the scores, especially education. However, high-executive semantic tasks alone do not allow us to clearly distinguish LLD from AD or aMCI, as both pathologies seem to have a largely dialectical influential relationship, but low-executive semantic tasks appear as more sensible to this pathological distinction.

Case Report: Intercurrent infections in COVID-19-induced sustained immunodepression: is interferon gamma a suitable drug?

Acute immuno-depression syndrome (AIDs) had been observed in many life-threatening conditions leading to the Intensive Care Unit. and is associated with recurrent secondary infections. We report one COVID-19 patient with a severe ARDS, demonstrating acute immunodepression syndrome lasting for several weeks. The occurrence of secondary infections despite long treatment by antibiotics led to combined interferon γ (IFNγ) as reported previously. The response to IFNγ was evaluated by the flowcytometry HLA-DR expression on circulating monocytes, which was repeated from time to time. The severe COVID-19 patients responded well to IFNγ without adverse events.

Barotrauma in COVID-19 acute respiratory distress syndrome: retrospective analysis of the COVADIS prospective multicenter observational database.

BACKGROUND: Despite evidence suggesting a higher risk of barotrauma during COVID-19-related acute respiratory distress syndrome (ARDS) compared to ARDS due to other causes, data are limited about possible associations with patient characteristics, ventilation strategy, and survival. METHODS: This prospective observational multicenter study included consecutive patients with moderate-to-severe COVID-19 ARDS requiring invasive mechanical ventilation and managed at any of 12 centers in France and Belgium between March and December 2020. The primary objective was to determine whether barotrauma was associated with ICU mortality (censored on day 90), and the secondary objective was to identify factors associated with barotrauma. RESULTS: Of 586 patients, 48 (8.2%) experienced barotrauma, including 35 with pneumothorax, 23 with pneumomediastinum, 1 with pneumoperitoneum, and 6 with subcutaneous emphysema. Median time from mechanical ventilation initiation to barotrauma detection was 3 [0-17] days. All patients received protective ventilation and nearly half (23/48) were in volume-controlled mode. Barotrauma was associated with higher hospital mortality (P < 0.001) even after adjustment on age, sex, comorbidities, PaO2/FiO2 at intubation, plateau pressure at intubation, and center (P < 0.05). The group with barotrauma had a lower mean body mass index (28.6 ± 5.8 vs. 30.3 ± 5.9, P = 0.03) and a higher proportion of patients given corticosteroids (87.5% vs. 63.4%, P = 0.001). CONCLUSION: Barotrauma during mechanical ventilation for COVID-19 ARDS was associated with higher hospital mortality.

Executive Profile of the Logopenic Variant of Primary Progressive Aphasia: Comparison with the Semantic and Non-Fluent Variants and Alzheimer's Disease.

The logopenic variant of primary progressive aphasia (lvPPA) shows different features from the non-fluent (nfvPPA) and semantic (svPPA) variants of PPA. Although language impairments remain the core symptoms, studies have highlighted the presence of executive disorders at the onset of the disease. Nevertheless, the results are contradictory, particularly in lvPPA. The aim of this study was to explore the executive profile of lvPPA. We compared executive functioning in lvPPA with the other two variants of PPA, Alzheimer's disease (AD) and a cognitively healthy group. In total, 70 patients with PPA, 32 patients with AD, and 41 healthy controls were included. They underwent a comprehensive executive battery assessing short-term and working memory, inhibition, flexibility, planning, and initiation. The analyses showed significant differences between the lvPPA group and the control group, except on visuospatial spans and the Stroop test, and between the lvPPA group and the other PPA groups and the AD group for several tasks. Thus, this research highlighted the existence of an executive dysfunction from the onset of the disease in lvPPA but also in the other two variants of PPA.

Physiological reactivity at rest and in response to social or emotional stimuli after a traumatic brain injury: A systematic review.

Numerous studies have shown that alterations in physiological reactivity (PR) after traumatic brain injury (TBI) are possibly associated with emotional deficits. We conducted a systematic review of these studies that evaluated PR in adults with moderate-to-severe TBI, either at rest or in response to emotional, stressful, or social stimuli. We focused on the most common measures of physiological response, including heart rate (HR), heart rate variability (HRV), respiratory sinus arrhythmia (RSA), electrodermal activity (EDA), salivary cortisol, facial electromyography (EMG), and blink reflex. Methods: A systematic literature search was conducted across six databases (PsycINFO, Psycarticles, SciencDirect, Cochrane Library, PubMed, and Scopus). The search returned 286 articles and 18 studies met the inclusion criteria. Results: Discrepancies were observed according to the type of physiological measure. Reduced physiological responses in patients with TBI have been reported in most EDA studies, which were also overrepresented in the review. In terms of facial EMG, patients with TBI appear to exhibit reduced activity of the corrugator muscle and diminished blink reflex, while in most studies, zygomaticus contraction did not show significant differences between TBI and controls. Interestingly, most studies measuring cardiac activity did not find significant differences between TBI and controls. Finally, one study measured salivary cortisol levels and reported no difference between patients with TBI and controls. Conclusion: Although disturbed EDA responses were frequently reported in patients with TBI, other measures did not consistently indicate an impairment in PR. These discrepancies could be due to the lesion pattern resulting from TBI, which could affect the PR to aversive stimuli. In addition, methodological differences concerning the measurements and their standardization as well as the characteristics of the patients may also be involved in these discrepancies. We propose methodological recommendations for the use of multiple and simultaneous PR measurements and standardization. Future research should converge toward a common methodology in terms of physiological data analysis to improve inter-study comparisons.

Electrophysiological correlates of attentional bias towards threat in children with high levels of social anxiety.

Pediatric social anxiety is characterized by attentional biases (AB) towards social threats. This study used a new response-based calculation method to assess AB from response times (RT) in a visual dot-probe task and electroencephalography (EEG) to explore its electrophysiological correlates. Twenty, high socially anxious children (HSA) (mean [M ] = 10.1 years; standard deviation [SD] = 1.01) were compared with 22 healthy control children (HC) (M = 10.20 years; SD = 1.30) matched in age and gender. Participants had to identify targets preceded by disgust-neutral, happy-neutral, or neutral-neutral pairs of faces. RT and electroencephalograms were recorded throughout the task. While no significant group difference was found at the behavioral level, principal component analyses performed on EEG data revealed that event-related potentials for threat-related stimuli were impacted by social anxiety. Analyses indicated a larger N170 amplitude in response to all facial stimuli in HC when compared to the HSA. However, we found increased P2 amplitudes for disgust-neutral pairs compared with happy-neutral pairs in has only. Then, thasHSA group showed increased P2 amplitudes for targets following disgusted faces on the opposite side of the screen compared with targets appearing on the same side of the screen. These results suggest that HSA may display an increased anchorage of attention on threatening stimuli and need more effort to disengage their attentional focus from threats and to perform the task correctly. Taken together, our data confirmed the presence of AB in children with high levels of social anxiety, which are reflected by increased neural processing during the confrontation to faces depicting a potential threatening expression.

Biases in BCI experiments: Do we really need to balance stimulus properties across categories?

Brain Computer Interfaces (BCIs) consist of an interaction between humans and computers with a specific mean of communication, such as voice, gestures, or even brain signals that are usually recorded by an Electroencephalogram (EEG). To ensure an optimal interaction, the BCI algorithm typically involves the classification of the input signals into predefined task-specific categories. However, a recurrent problem is that the classifier can easily be biased by uncontrolled experimental conditions, namely covariates, that are unbalanced across the categories. This issue led to the current solution of forcing the balance of these covariates across the different categories which is time consuming and drastically decreases the dataset diversity. The purpose of this research is to evaluate the need for this forced balance in BCI experiments involving EEG data. A typical design of neural BCIs involves repeated experimental trials using visual stimuli to trigger the so-called Event-Related Potential (ERP). The classifier is expected to learn spatio-temporal patterns specific to categories rather than patterns related to uncontrolled stimulus properties, such as psycho-linguistic variables (e.g., phoneme number, familiarity, and age of acquisition) and image properties (e.g., contrast, compactness, and homogeneity). The challenges are then to know how biased the decision is, which features affect the classification the most, which part of the signal is impacted, and what is the probability to perform neural categorization per se. To address these problems, this research has two main objectives: (1) modeling and quantifying the covariate effects to identify spatio-temporal regions of the EEG allowing maximal classification performance while minimizing the biasing effect, and (2) evaluating the need to balance the covariates across categories when studying brain mechanisms. To solve the modeling problem, we propose using a linear parametric analysis applied to some observable and commonly studied covariates to them. The biasing effect is quantified by comparing the regions highly influenced by the covariates with the regions of high categorical contrast, i.e., parts of the ERP allowing a reliable classification. The need to balance the stimulus's inner properties across categories is evaluated by assessing the separability between category-related and covariate-related evoked responses. The procedure is applied to a visual priming experiment where the images represent items belonging to living or non-living entities. The observed covariates are the commonly controlled psycho-linguistic variables and some visual features of the images. As a result, we identified that the category of the stimulus mostly affects the late evoked response. The covariates, when not modeled, have a biasing effect on the classification, essentially in the early evoked response. This effect increases with the diversity of the dataset and the complexity of the algorithm used. As the effects of both psycho-linguistic variables and image features appear outside of the spatio-temporal regions of significant categorical contrast, the proper selection of the region of interest makes the classification reliable. Having proved that the covariate effects can be separated from the categorical effect, our framework can be further used to isolate the category-dependent evoked response from the rest of the EEG to study neural processes involved when seeing living vs. non-living entities.

Septic Shock: Phenotypes and Outcomes.

INTRODUCTION: Sepsis is a heterogeneous syndrome that results in life-threatening organ dysfunction. Our goal was to determine the relevant variables and patient phenotypes to use in predicting sepsis outcomes. METHODS: We performed an ancillary study concerning 119 patients with septic shock at intensive care unit (ICU) admittance (T0). We defined clinical worsening as having an increased sequential organ failure assessment (SOFA) score of ≥ 1, 48 h after admission (ΔSOFA ≥ 1). We performed univariate and multivariate analyses based on the 28-day mortality rate and ΔSOFA ≥ 1 and determined three patient phenotypes: safe, intermediate and unsafe. The persistence of the intermediate and unsafe phenotypes after T0 was defined as a poor outcome. RESULTS: At T0, the multivariate analysis showed two variables associated with 28-day mortality rate: norepinephrine dose and serum lactate concentration. Regarding ΔSOFA ≥ 1, we identified three variables at T0: norepinephrine dose, lactate concentration and venous-to-arterial carbon dioxide difference (P(v-a)CO2). At T0, the three phenotypes (safe, intermediate and unsafe) were found in 28 (24%), 70 (59%) and 21 (18%) patients, respectively. We thus suggested using an algorithm featuring norepinephrine dose, lactate concentration and P(v-a)CO2 to predict patient outcomes and obtained an area under the curve (AUC) of 74% (63-85%). CONCLUSION: Our findings highlight the fact that identifying relevant variables and phenotypes may help physicians predict patient outcomes.

Normative Data for Healthy French-Speaking Persons Aged 80 Years and Older for the DTLA Language Screening Test.

The Detection Test for Language Impairments in Adults and the Aged (DTLA) is a quick, sensitive, and standardized screening test designed to assess language disorders in adults and elderly people. The test was specifically developed to detect linguistic impairment associated with major neurocognitive disorders. In 2017, we established normative data on 545 healthy individuals between 50 and 80 years old from four French-speaking countries: Belgium, Canada (Quebec), France, and Switzerland. OBJECTIVE: The objective of the present study was to develop norms for the population older than 80 years of age for the DTLA. METHOD: We extend the original normative data to include 149 healthy, community-dwelling, French-speaking adults aged 80 years old and older from the same countries. RESULTS: For the total score of the screening test, we calculated the 5th, 15th, 25th, and 50th percentiles for two education groups. The analyses allowed the identification of cutoff and alert scores based on education level. CONCLUSIONS: With the present study, solid normative data for the DTLA derived from the performance of 694 healthy, community-dwelling adults, and elderly people are now available to clinicians and researchers.

French Adaptation of the Brief Irritability Test: Factor Structure, Psychometric Properties, and Relationship with Depressive Symptoms.

The Brief Irritability Test (BITe, Holtzman et al., 2014) is a brief, reliable, and valid self-report measure of irritability. Despite the growing interest to understand the underlying causes and consequences of irritability, this questionnaire has not been developed and validated for a French-speaking population yet. In the present study, 413 participants completed our French adaptation of the BITe (i.e., TCI; Test Court d'Irritabilité) and measures of associated constructs (depression, anger, hostility, and aggression) and well-being (life satisfaction and social support). Descriptive, psychometric (i.e., Cronbach alpha and Spearman correlation coefficients), and factor analyses were conducted. An exploratory factor analysis in sample 1 (n = 209), yielded one single factor. The confirmatory factor analysis in sample 2 (n = 204) showed a reasonable fit of this single factor model explaining 55.5% of the variance and presenting a strong internal consistency (α = .80). Compared to the original English questionnaire, the TCI shares similar unidimensional factor organization and correlations with other constructs, although a gender bias was identified, with women scoring higher than men. Irritability was higher among respondents in the age range 17-25, compared to older adults. A hierarchical regression analysis showed that TCI scores significantly predict depressive symptoms when demographics were controlled for. In summary, the TCI presents good psychometric properties and could constitute a valuable tool to evaluate irritability in clinical and research contexts.

Neural responses associated with attentional engagement and disengagement from threat in high socially anxious children: Evidence from temporal-spatial PCA.

BACKGROUND: Cognitive models indicated that social anxiety disorder (SAD) would be caused and maintained by a biased attentional processing of threatening information. This study investigates whether socially anxious children may present impaired attentional engagement and disengagement from negative emotional faces, as well as their underlying event-related potential responses. METHODS AND FINDINGS: Fifteen children with high levels of social anxiety (HSA; 9 boys; mean age = 9.99y; SD = 1.14) and twenty low socially anxious children (LSA; 16 boys; mean age = 10.47y; SD = 1.17) participated in a spatial cueing task in which they had to detect targets following neutral/disgusted faces in a valid or invalid location. No group effect was reported on reaction times [p>.05]. However, electrophysiological data showed lower P3a amplitude in HSA children compared with the LSA group when processing facial stimuli. They also reported larger N2 amplitudes for valid-disgusted targets and a larger P3a amplitude for the invalid-disgusted ones. CONCLUSION: In terms of electrophysiological data, our results validated, the hypothesis of attentional disengagement difficulties in SAD children. We also confirm the idea that high levels of social anxiety are associated with cognitive control impairments and have a greater impact on the processing efficiency than on the performance effectiveness.

Covert Attention to Gestures Is Sufficient for Information Uptake.

Numerous studies have explored the benefit of iconic gestures in speech comprehension. However, only few studies have investigated how visual attention was allocated to these gestures in the context of clear versus degraded speech and the way information is extracted for enhancing comprehension. This study aimed to explore the effect of iconic gestures on comprehension and whether fixating the gesture is required for information extraction. Four types of gestures (i.e., semantically and syntactically incongruent iconic gestures, meaningless configurations, and congruent iconic gestures) were presented in a sentence context in three different listening conditions (i.e., clear, partly degraded or fully degraded speech). Using eye tracking technology, participants' gaze was recorded, while they watched video clips after which they were invited to answer simple comprehension questions. Results first showed that different types of gestures differently attract attention and that the more speech was degraded, the less participants would pay attention to gestures. Furthermore, semantically incongruent gestures appeared to particularly impair comprehension although not being fixated while congruent gestures appeared to improve comprehension despite also not being fixated. These results suggest that covert attention is sufficient to convey information that will be processed by the listener.

Late-Life Depression, Cognitive Impairment, and Relationship with Alzheimer's Disease.

This narrative review aimed to explore the existing knowledge in order to examine the multiple forms of late-life depression (LLD) within a non-neurodegenerative or a neurodegenerative context, in particular Alzheimer's disease (AD). This review will first provide information about different pathogenic hypotheses proposed to describe LLD when it is not linked to a neurodegenerative context. Within the presentation of these syndromes, the literature reports thymic and cognitive specific features and highlights a common preponderance of cognitive impairment, and particularly executive. This review will also report data from research works that have addressed the role of depressive symptoms (DSs) in incidence of AD and mild cognitive impairment (MCI) conversion to AD. These findings support the claim that the relationship between DS (or LLD) and the cognitive decline encountered in AD can be of 2 types: (1) risk factor or (2) prodrome. They also support the hypothesis that the effect of DS on the incidence of AD can be identified between specific characteristics of these symptoms such as a very first apparition late in life, an increasing severity of DS, and a poor response to medical treatment. Finally, longitudinal and cross-sectional research will be presented, aiming to identify the predictive value of DS (or LLD) on AD incidence and/or conversion of MCI (and specifically amnestic MCI). This final section shows that specific features of LLD, such as being of early- or late-onset, can be considered as indices of AD incidence.

Rehabilitation of language processing disorders in bilingual patients with Alzheimer's disease.

We studied language information processing abilities in subjects with Alzheimer's disease (AD) who were bilingual (French and dialectical Arabic). The results show a disruption of certain semantic aspects of their mother tongue (L1); a process based on the lexical implications of terms and continuity. On the other hand, grammatical levels appeared to be relatively unaffected in oral speech in L1 but were disturbed in the second language (L2). We consequently adapted a cognitive-language stimulation protocol for bilingual patients (PSCLAB) to respond to these deficits. The effectiveness of the PSCLAB in terms of rehabilitation was assessed in 30 such patients through discourse analysis carried out before and after initiating the protocol. The results show that cognitive/language training using the PSCLAB appears to improve language behaviour of bilingual patients with AD, although this should be confirmed through larger controlled studies.

Influence of Psychological Vulnerability Factors for Bipolar Disorders on a Semantic Mediated Priming Task.

Hypomanic personality, hyperthymic temperament and irritable temperament are considered as psychological vulnerability factors to bipolar disorders. Semantic memory is impaired in bipolar patients. Spreading activation is among the probable candidates for accounting this impairment. The aim of this study was to assess spreading activation according to vulnerability factors continuum to determine whether it could be a factor of vulnerability to bipolar disorders. A sample of 61 healthy volunteers was recruited. Spreading activation was assessed by semantic mediated priming implemented in a double lexical decision task. Results shown that semantic mediated priming was negatively associated to hyperthymic temperament and irritable temperament. Impairment in semantic memory, and more specifically spreading activation, appear to be a cognitive factor of vulnerability to bipolar disorders. Our results can contribute to a better understanding of semantic impairment in vulnerable population and in bipolar disorder.

Perceptual and Interoceptive Strength Norms for 270 French Words.

Perceptual experience through the five modalities (i.e., vision, hearing, touch, taste, and smell) has demonstrated its key role in semantics. Researchers also highlighted the role of interoceptive information in the grounded representation of concepts. However, to this day, there is no available data across these modalities in the French language. Therefore, the aim of this study was to circumvent this caveat. Participants aged between 18 and 50 completed an online survey in which we recorded scores of perceptual strength (PS), interoceptive information, imageability, concreteness, conceptual familiarity, and age of acquisition of 270 words of the French language. We also analysed the relationships between perceptual modalities and psycholinguistic variables. Results showed that vast majority of concepts were visually-dominant. Correlation analyses revealed that the five PS variables were strongly correlated with imageability, concreteness, and conceptual familiarity and highlight that PS variables index one aspect of the semantic representations of a word. On the other hand, high interoceptive scores were highlighted only for the less imageable and less concrete words, emphasizing their importance for the grounding of abstract concepts. Future research could use these norms in the investigation of the role of perceptual experience in the representation of concepts and their impact on word processing.

Eye-tracking exploration of inhibitory control in post-traumatic stress disorder: an emotional antisaccade paradigm.

Background: Cognitive-behavioural studies among individuals suffering from post-traumatic stress disorder (PTSD) have highlighted attentional biases towards threats as a key factor in the maintenance of the disorder. Anxiety-related studies have hypothesized that attentional biases were due to attentional control difficulties in inhibition and flexibility of threatening information. Objective: Because it remains unclear how this theory could be applied to PTSD, this study aims to evaluate the inhibitory control and flexibility abilities of negative and threatening information in this population, using eye-tracking technology. Method: Fifteen adults with a history of physical assault and a current diagnosis of PTSD, and 15 healthy control participants, completed an original mixed antisaccade task. Results: We found enhanced overt attentional allocation towards every item of emotional information among PTSD participants, such as indexed by the latencies of the first saccade in prosaccade trials, followed by disengagement difficulties, such as indexed by increased reaction time to identify the target. Conclusion: Our results could represent empirical evidence of the general enhancement of attentional vigilance in people with PTSD in comparison with healthy controls, as well as specific inhibitory deficits. The results are interpreted through a fear-generalization hypothesis.

Antecedentes: Los estudios cognitivo-conductuales entre personas que padecen Trastorno de Estrés Postraumático (TEPT) han destacado los sesgos atencionales (AB, por su sigla en inglés) por amenaza como factor clave para el mantenimiento del trastorno. La literatura relacionada con ansiedad ha planteado la hipótesis de que los AB se debían a la dificultad del control atencional en la inhibición y flexibilidad de la información amenazante (Eysenck, 2008).Objetivo: Debido a que no está claro cómo se podría aplicar esta teoría al TEPT, este estudio tiene como objetivo evaluar el control inhibitorio y las capacidades de flexibilidad de la información negativa y amenazante en esta población, utilizando tecnología de seguimiento-ocular.Método: 15 adultos agredidos físicamente con un diagnóstico actual de TEPT y 15 participantes de Controles Sanos (CS) completaron una tarea original de antisacada mixta.Resultados: Encontramos una asignación atencional directa-abierta aumentada hacia información emocional entre los participantes con TEPT, tales como las latencias indexadas de la primera sacada en los ensayos de prosacada, seguida por las dificultades de desenganche, indexada por un tiempo de reacción mayor para identificar el objetivo diana.Conclusión: Nuestros resultados podrían representar una evidencia empírica de un aumento general de la vigilancia atencional en el TEPT en comparación con los CS, así como de déficits inhibitorios específicos. Los resultados se interpretan a través de una hipótesis de generalización del miedo.